Facial-Landmark-Detection: Optimized for Mobile Deployment

Real-time 3D facial landmark detection optimized for mobile and edge

Detects facial landmarks (eg, nose, mouth, etc.). This model's architecture was developed by Qualcomm. The model was trained by Qualcomm on a proprietary dataset of faces, but can be used on any image.

This repository provides scripts to run Facial-Landmark-Detection on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

• Model Type: Model_use_case.pose_estimation
• Model Stats:
  • Input resolution: 128x128
  • Number of parameters: 5.42M
  • Model size (float): 20.7 MB
  • Model size (w8a8): 5.27 MB

Model	Precision	Device	Chipset	Target Runtime	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit	Target Model
Facial-Landmark-Detection	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	1.131 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	1.137 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	0.481 ms	0 - 146 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	0.551 ms	0 - 130 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	0.281 ms	0 - 10 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	0.298 ms	0 - 3 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	0.514 ms	0 - 15 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	0.503 ms	0 - 116 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	0.515 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	SA7255P ADP	Qualcomm® SA7255P	TFLITE	1.131 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	1.137 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	TFLITE	0.284 ms	0 - 4 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	QNN_DLC	0.299 ms	0 - 3 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	SA8295P ADP	Qualcomm® SA8295P	TFLITE	0.641 ms	0 - 121 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	0.64 ms	0 - 120 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	TFLITE	0.283 ms	0 - 2 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	QNN_DLC	0.302 ms	0 - 2 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	SA8775P ADP	Qualcomm® SA8775P	TFLITE	0.503 ms	0 - 116 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	0.515 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	0.225 ms	0 - 144 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	0.236 ms	0 - 129 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	0.332 ms	0 - 103 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	0.19 ms	0 - 120 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	0.198 ms	0 - 118 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	0.318 ms	0 - 90 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	TFLITE	0.189 ms	0 - 119 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_DLC	0.196 ms	0 - 118 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	ONNX	0.316 ms	0 - 91 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	float	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	0.367 ms	0 - 0 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	float	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	0.394 ms	10 - 10 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Dragonwing Q-6690 MTP	Qualcomm® Qcm6690	TFLITE	0.604 ms	0 - 123 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Dragonwing Q-6690 MTP	Qualcomm® Qcm6690	QNN_DLC	0.597 ms	0 - 124 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Dragonwing Q-6690 MTP	Qualcomm® Qcm6690	ONNX	1.664 ms	0 - 13 MB	CPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	TFLITE	0.604 ms	0 - 7 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	QNN_DLC	0.693 ms	0 - 2 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Dragonwing RB3 Gen 2 Vision Kit	Qualcomm® QCS6490	ONNX	2.776 ms	2 - 12 MB	CPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	TFLITE	0.46 ms	0 - 114 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	QCS8275 (Proxy)	Qualcomm® QCS8275 (Proxy)	QNN_DLC	0.43 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	TFLITE	0.27 ms	0 - 138 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	QCS8450 (Proxy)	Qualcomm® QCS8450 (Proxy)	QNN_DLC	0.247 ms	0 - 137 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	TFLITE	0.169 ms	0 - 3 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	QNN_DLC	0.17 ms	0 - 2 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	QCS8550 (Proxy)	Qualcomm® QCS8550 (Proxy)	ONNX	0.365 ms	0 - 9 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	TFLITE	0.331 ms	0 - 114 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	QCS9075 (Proxy)	Qualcomm® QCS9075 (Proxy)	QNN_DLC	0.311 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	RB5 (Proxy)	Qualcomm® QCS8250 (Proxy)	TFLITE	3.655 ms	0 - 37 MB	GPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	RB5 (Proxy)	Qualcomm® QCS8250 (Proxy)	ONNX	1.537 ms	0 - 18 MB	CPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	SA7255P ADP	Qualcomm® SA7255P	TFLITE	0.46 ms	0 - 114 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	SA7255P ADP	Qualcomm® SA7255P	QNN_DLC	0.43 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	TFLITE	0.176 ms	0 - 4 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	SA8255 (Proxy)	Qualcomm® SA8255P (Proxy)	QNN_DLC	0.165 ms	0 - 3 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	SA8295P ADP	Qualcomm® SA8295P	TFLITE	0.441 ms	0 - 120 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	SA8295P ADP	Qualcomm® SA8295P	QNN_DLC	0.443 ms	0 - 121 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	TFLITE	0.188 ms	0 - 3 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	SA8650 (Proxy)	Qualcomm® SA8650P (Proxy)	QNN_DLC	0.169 ms	0 - 2 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	SA8775P ADP	Qualcomm® SA8775P	TFLITE	0.331 ms	0 - 114 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	SA8775P ADP	Qualcomm® SA8775P	QNN_DLC	0.311 ms	0 - 115 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	TFLITE	0.14 ms	0 - 139 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	QNN_DLC	0.14 ms	0 - 139 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Samsung Galaxy S24	Snapdragon® 8 Gen 3 Mobile	ONNX	0.227 ms	0 - 114 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	TFLITE	0.123 ms	0 - 118 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	QNN_DLC	0.118 ms	0 - 119 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Samsung Galaxy S25	Snapdragon® 8 Elite For Galaxy Mobile	ONNX	0.209 ms	0 - 91 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	TFLITE	0.224 ms	0 - 123 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	QNN_DLC	0.22 ms	0 - 124 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Snapdragon 7 Gen 4 QRD	Snapdragon® 7 Gen 4 Mobile	ONNX	1.553 ms	2 - 18 MB	CPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	TFLITE	0.126 ms	0 - 116 MB	NPU	Facial-Landmark-Detection.tflite
Facial-Landmark-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	QNN_DLC	0.119 ms	0 - 117 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Snapdragon 8 Elite Gen 5 QRD	Snapdragon® 8 Elite Gen5 Mobile	ONNX	0.27 ms	0 - 93 MB	NPU	Facial-Landmark-Detection.onnx.zip
Facial-Landmark-Detection	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	QNN_DLC	0.236 ms	0 - 0 MB	NPU	Facial-Landmark-Detection.dlc
Facial-Landmark-Detection	w8a8	Snapdragon X Elite CRD	Snapdragon® X Elite	ONNX	0.232 ms	5 - 5 MB	NPU	Facial-Landmark-Detection.onnx.zip

Installation

Install the package via pip:

# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[facemap-3dmm]"

Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub Workbench with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.facemap_3dmm.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.facemap_3dmm.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

• Performance check on-device on a cloud-hosted device
• Downloads compiled assets that can be deployed on-device for Android.
• Accuracy check between PyTorch and on-device outputs.

python -m qai_hub_models.models.facemap_3dmm.export

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.facemap_3dmm import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)
        

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub Workbench. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.facemap_3dmm.demo --eval-mode on-device

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.facemap_3dmm.demo -- --eval-mode on-device

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

• TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.

• QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on Facial-Landmark-Detection's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

• The license for the original implementation of Facial-Landmark-Detection can be found here.

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.